General Session 3

M. Lambertini¹, E. Blondeaux², V. Fontana², S. Cinieri³, I. Paris⁴, A. Vandone⁵, C. Andreetta⁶, V. Sini⁷, I. Meattini⁸, C. Bengala⁹, M. Dieci¹⁰, L. Livraghi¹¹, R. Cioffi¹², D. Bilancia¹³, F. Sottotetti¹⁴, S. Mura¹⁵, A. Denaro¹⁶, D. Favero¹⁷, G. Arpino¹⁸, F. Poggio¹⁹, M. Razeti¹⁹, E. Chiappe¹⁹, F. Bruzzone¹⁹, L. Arecco²⁰, C. Massarotti²¹, P. Vanella⁵, F. Pavese²², L. Orlando²³, F. Puglisi²⁴, P. Anserini²¹, L. Boni², L. Del Mastro¹⁹; ¹Medical Oncology, University of Genova – IRCCS Ospedale Policlinico San Martino, Genoa, ITALY, ²Clinical Trial Unit, University of Genova – IRCCS Ospedale Policlinico San Martino, Genova, ITALY, ³Medical Oncology, Medical Oncology Department and Breast Unit, Perrino Hospital, Brindisi, ITALY, ⁴Medical Oncology, Division of Gynecologic Oncology, Department of Woman and Child Health and Public Health, Fondazione Policlinico Universitario Agostino Gemelli IRCCS, Rome, ITALY, ⁵Medical Oncology, Azienda Ospedaliera Santa Croce e Carle, Cuneo, ITALY, ⁶Medical Oncology, Azienda Sanitaria Universitaria Friuli Centrale, Udine, ITALY, ⁷Medical Oncology, Centro Oncologico Santo Spirito-Nuovo Regina Margherita, ASL Roma 1, Rome, ITALY, ⁸Oncology, Oncology Department – Florence University Hospital, Florence, ITALY, ⁹Medical Oncology, Azienda Ospedaliero-Universitaria Pisana, Pisa, ITALY, ¹⁰Medical Oncology, Istituto Oncologico Veneto IOV-IRCCS, Padova, ITALY, ¹¹Medical Oncology, Hospital of Prato, Azienda USL Toscana Centro, Prato, ITALY, ¹²Obstetrics and Gynecology, IRCCS San Raffaele Hospital, Milan, ITALY, ¹³Medical Oncology, Azienda Ospedaliera S. Carlo, Potenza, ITALY, ¹⁴Medical Oncology, ICS Maugeri IRCCS, Pavia, ITALY, ¹⁵Medical Oncology, AOU Ospedale SS Annunziata, Sassari, ITALY, ¹⁶Medical Oncology, Ospedale Maggiore di Trieste, Trieste, ITALY, ¹⁷Medical Oncology, Candiolo Cancer Institute, FPO-IRCCS, Torino, ITALY, ¹⁸Medical Oncology, University of Naples “Federico II”, Naples, ITALY, ¹⁹Medical Oncology, University of Genova – IRCCS Ospedale Policlinico San Martino, Genova, ITALY, ²⁰Medical Oncology, Université libre de Bruxelles (ULB), Hôpital Universitaire de Bruxelles (H.U.B), Clinical Trials Unit, Institut Jules Bordet, Brussels, BELGIUM, ²¹Obstetrics and Gynecology, University of Genova – IRCCS Ospedale Policlinico San Martino, Genova, ITALY, ²²Medical Oncology, Department of Woman and Child Health and Public Health, Fondazione Policlinico Universitario Agostino Gemelli IRCCS, Rome, ITALY, ²³Medical Oncology, Perrino Hospital, Brindisi, ITALY, ²⁴Medical Oncology, Centro di Riferimento Oncologico di Avano (CRO) IRCCS, Aviano, ITALY.

Background: Oncofertility counseling is standard of care in premenopausal patients (pts) with early breast cancer (EBC) who are candidates to receive chemotherapy (CT). However, limited data exist on the uptake of ovarian function and/or fertility preservation strategies and reasons for refusal. Moreover, for pts undergoing controlled ovarian stimulation (COS) for gamete cryopreservation, the evidence on its safety derives mostly from retrospective studies; thus, some physicians remain concerned in offering this strategy. To address the significant challenges of the oncofertility counseling in premenopausal women with EBC, the PREFER study was developed as a national comprehensive program aiming to optimize care and improve knowledge in this field. Methods: PREFER (NCT02895165) is a multicenter prospective cohort study conducted in 23 Italian centers affiliated to the Gruppo Italiano Mammella (GIM) study group. Eligible pts were premenopausal women aged <=45 years at diagnosis of EBC. Before starting (neo)adjuvant CT, a proper oncofertility counseling was offered. Pts aged <=40 years were offered oocyte cryopreservation (OC; embryo cryopreservation is forbidden by law in Italy) and/or ovarian tissue cryopreservation (OTC) for preserving fertility, as well as the use of a gonadotropin-releasing hormone agonist (GnRHa) during CT for ovarian function preservation. Pts aged 41-45 years were only offered the use of a GnRHa during CT for ovarian function preservation. Universal coverage of all these procedures is provided by the Italian National Health System. The primary objective was to assess the acceptance rates of these strategies and reasons for refusal. Assessing the safety of COS for OC in terms of disease-free survival (DFS) and overall survival (OS) was a secondary objective. Results: Between November 2012 and December 2024, 746 pts were enrolled, of whom 533 were aged <=40 years and 213 between 41-45 years. Median age was 38 years (IQR 34-41 years); 537 (72.0%) pts had hormone receptor-positive EBC and 93 (12.5%) were germline BRCA carriers. Among pts aged 41-45 years, 160 (75.1%) accepted use of a GnRHa during CT. Main reasons for refusal were completion of family planning in 24 (11.3 %) pts. Among pts aged <=40 years, 476 (89.3%) accepted use of a GnRHa during CT. Main reason for refusal was completion of family planning in 18 (3.4%) pts. A cryopreservation technique was accepted by 133 (25.0%) pts, of whom 127 (23.8%) underwent OC and 6 (1.1%) OTC. Main factors associated with acceptance of these strategies were younger age at diagnosis (OR 0.80; 95% CI 0.76-0.85 for 1-year increase), no prior children (OR 0.08; 95% CI 0.04-0.13), more recent year at EBC diagnosis (OR 1.37; 95% CI 0.74-2.51 for 2016-2020 vs. 2012-2015; OR 2.06; 95% CI 1.15-3.67 for 2021-2024 vs. 2012-2015) and lower tumor grade (OR 0.55; 95% CI 0.36-0.85 for G3 vs. G1-2). No difference in acceptance of these strategies according to hormone receptor status, BRCA status, timing of CT administration, tumor size and nodal status was observed. Median follow-up was 3.8 years (IQR 1.7-6.6 years). Undergoing COS for OC was not associated with any detrimental effect on DFS (adjusted HR 0.78; 95% CI 0.27-2.26) nor on OS (adjusted HR 0.53; 95% CI 0.06-4.73). Conclusions: PREFER is the largest multicenter prospective cohort study addressing oncofertility care in premenopausal women with EBC. Most pts (89.3% of those <=40 years and 75.1% of those 41-45 years) accepted a GnRHa during CT for ovarian function preservation. One out of four (25.0%) pts aged <=40 years accepted to undergo cryopreservation procedures (mainly OC). COS for OC did not appear to negatively influence prognosis. These results provide novel prospective evidence to improve the oncofertility counseling of premenopausal women with EBC.

E. P. Mamounas¹, H. Bandos², K. J. Sweeney³, K. M. Tran⁴, E. Chen⁵, P. Rastogi⁶, V. Valero⁷, T. J. Freeman⁸, C. E. Geyer, Jr⁹, L. Fehrenbacher¹⁰, S. K. Chia¹¹, A. M. Brufsky¹², J. M. Walshe¹³, G. S. Soori¹⁴, S. R. Dakhil¹⁵, S. Paik¹⁶, S. M. Swain¹⁷, N. Wolmark¹⁸, W. F. Symmans¹⁹; ¹Radiation Oncology, AdventHealth Cancer Institute, Orlando, FL, ²Department of Biostatistics and Health Data Science, NRG Oncology Statistical and Data Management Center, University of Pittsburgh, Pittsburgh, PA, ³Department of Anatomical Pathology, Division of Pathology-Lab Medicine Div, University of Texas MD Anderson Cancer Center, Houston, TX, ⁴Lab, University of Texas MD Anderson Cancer Center, Houston, TX, ⁵Lab, The University of Texas MD Anderson Cancer Center, Houston, TX, ⁶Hematology and Medical Oncology, NSABP Foundation, Inc.; UPMC Hillman Cancer Center; University of Pittsburgh School of Medicine, Pittsburgh, PA, ⁷Breast Medical Oncology, The University of Texas MD Anderson Cancer Center, Houston, TX, ⁸Department of Pathology, NSABP Foundation, Inc.; University of Pittsburgh School of Medicine, Pittsburgh, PA, ⁹Malignant Hematology and Medical Oncology, NSABP Foundation, Inc.; UPMC Hillman Cancer Center; University of Pittsburgh School of Medicine, Pittsburgh, PA, ¹⁰Radiation Oncology, Kaiser Permanente Oncology Clinical Trials Northern CA, Vallejo, CA, ¹¹Medical Oncology, British Columbia Cancer Agency; Division of Medical Oncology, UBC Department of Medicine; BC Cancer – Vancouver Cancer Centre, Vancouver, BC, CANADA, ¹²Division of Hematology-Oncology, UPMC Hillman Cancer Center; University of Pittsburgh School of Medicine; Magee-Womens Hospital, Pittsburgh, PA, ¹³Medical Oncology, St. Vincent’s University Hospital and Cancer Trials Ireland, Dublin, IRELAND, ¹⁴Medical Oncology, Florida Cancer Specialists, Ft. Meyers, FL, ¹⁵Medical Oncology, Wichita NCORP, Via Christi Regional Medical Center; Cancer Center of Kansas, Wichita, KS, ¹⁶Onc/Hemat, Theragenbio, Inc, Pankyo, South Korea; Severance Biomedical Science Institute at Yonsei University College of Medicine, Seoul, South Korea, Seoul, KOREA, REPUBLIC OF, ¹⁷Medical Oncology, Georgetown Lombardi Comprehensive Cancer Center; Georgetown University Medical Center; and MedStar Health, Washington, DC, DC, ¹⁸Div Surgical Oncology, Dept of Surgery, NSABP Foundation, Inc.; UPMC Hillman Cancer Center; University of Pittsburgh School of Medicine, Pittsburgh, PA, ¹⁹Breast Surgical Pathology, University of Texas MD Anderson Cancer Center, Houston, TX.

Background: SET_ER/PR index of sensitivity to endocrine therapy (ET) measures endocrine receptor-related transcription from fixed paraffin-embedded tissue and is highly reproducible within and between laboratories. SET_ER/PR index is correlated with receptor ligand binding activity and predicts early pharmacodynamic response to ET and survival outcomes following ET in palliative and adjuvant treatment settings. We evaluated whether SET_ER/PR index predicts benefit from extended letrozole therapy (ELT) in the NSABP B-42 trial, which evaluated ELT vs placebo in HR+ postmenopausal breast cancer patients (pts) who had received 5 yrs of adjuvant ET with aromatase inhibitor (AI), or tamoxifen followed by AI. Methods: SET_ER/PR index was measured using the QuantiGene Plex bead-based hybridization assay (ThermoFisher) using two unstained sections and an H&E-stained slide in B-42 HER2- samples. Cut points for analysis of SET_ER/PR index were pre-defined from the inter-quartile range (1.10 ≤SET_ER/PR index ≤2.10) and the mean SET_ER/PR index of 1.50 from a reference cohort of samples, representative of B-42 eligibility. Primary endpoint was breast cancer-free interval (BCFI). Stratified Cox proportional hazards models were used to calculate hazard ratio with 95% confidence interval (HR, 95%CI) with significance at p<0.05. Results: Breast cancer blocks with tumor tissue were available from 1,556 eligible, HER2- pts, with 1,489 (96%) producing SET_ER/PR results that passed quality control and were included in the study cohort. There were no statistically significant differences in pt or tumor characteristics between the study cohort and the rest of the HER2 negative B-42 pts. ELT reduced the 10-yr rate of BCFI events by 4.2% in the study cohort (HR 0.69, 95%CI 0.51-0.94, p=0.016). Among 1,489 pts, 59% had SET_ER/PR index in the range 1.10-2.10 with 42% measuring ≥1.50. No difference in the ELT effect was observed between groups of pts identified by the primary inter-quartile range cut point (1.10 ≤SET_ER/PR index ≤2.10 vs other). Benefit from ELT was statistically significant in pts with SET_ER/PR index ≥1.50 (HR 0.53, 95%CI 0.32-0.88, p=0.014), but not in those with SET_ER/PR index <1.50 (HR 0.82, 95%CI 0.56-1.20, p=0.31), without significant treatment-by- SET_ER/PR index interaction (p=.14). Overall, the 10-yr absolute BCFI benefit from ELT was 7.1% in pts with SET_ER/PR index ≥1.50 (Placebo: 14.9%, Letrozole: 7.8%) and 2.1% in those with SET_ER/PR index <1.50 (Placebo: 14.5%, Letrozole 12.4%). When evaluated by nodal status, node-positive pts with SET_ER/PR index ≥1.50, had a 10.5% absolute reduction in the 10-yr BCFI events (Placebo:19.2%, Letrozole: 8.7%; HR 0.52, 95%CI 0.26-1.06), whereas those with SET_ER/PR index <1.50, had a 3% absolute reduction in the 10-yr BCFI events (Placebo:20.1%, Letrozole: 17.1%; HR 0.84, 95%CI 0.50-1.40). Node-negative pts with SET_ER/PR index ≥1.50 had a 5.2% absolute reduction in the 10-yr BCFI events (Placebo: 12.4%, Letrozole: 7.2%; HR 0.55, 95%CI 0.27-1.11), whereas those with SET_ER/PR index <1.50, had a 1.9% absolute reduction in the 10-yr BCFI events (Placebo:10.8%, Letrozole: 8.9%; HR 0.80, 95%CI 0.44-1.43). When evaluating the performance of SET_ER/PR index as a continuous measure, increasing values of SET_ER/PR index were associated with greater relative benefit from ELT (treatment-by- SET_ER/PR index [per unit increase] interaction HR 0.71, 95%CI 0.43-1.18). Conclusions: Benefit from ELT was enriched in the subset of pts with SET_ER/PR index ≥1.50. This supports the hypothesis that longer duration of endocrine therapy is more effective for pts with highly endocrine-sensitive cancers. NCT00382070 NCI UH3CA276603; U10 CA180868, -180822, UG1CA189867; U24CA196067; Korea Health Technology R&D Project.

A. C. Watt¹, A. Ahn², C. Blyth², J. Dixon-Douglas³, K. Ambani², R. Coulson², M. Taylor², K. Chan², C. Dietrich², B. E. Russ², S. Ramm², C. A. Mahendra², K. Lu², N. Pires², E. Lim⁴, S. Chandarlapaty⁵, F. Andre³, S. Goel¹; ¹Cancer Biology and THerapeutics, Peter MacCallum Cancer Centre, Melbourne, AUSTRALIA, ²Cancer Biology and Therapeutics, Peter MacCallum Cancer Centre, Melbourne, AUSTRALIA, ³Medical Oncology, Institut Gustave Roussy, Paris, FRANCE, ⁴Cancer Research, Garvan Institute of Medical Research, Dalinghurst, AUSTRALIA, ⁵Human Oncology and Pathogenesis, Memorial Sloan Kettering Cancer Centre, New York, NY.

CDK4/6 inhibitors induce durable responses in estrogen receptor-positive (ER+) breast cancer, primarily via activation of the tumor suppressor Rb. While Rb is classically viewed as a repressor of E2F-dependent cell cycle genes, evidence from the pre-genomic era suggested broader roles in gene regulation. Here, we identify a previously unrecognized function of Rb as a transcriptional activator in luminal breast cancer, with key implications for endocrine sensitivity and resistance. Using CUT&RUN, we mapped genome-wide Rb binding in ER+ breast cancer cell lines (MCF7, ZR-75-1) and a patient-derived xenograft (PDX) following abemaciclib or palbociclib treatment. CDK4/6 inhibition led to Rb hypo-phosphorylation and a large increase in chromatin association. As expected, Rb localized to E2F-regulated promoters, repressing canonical cell cycle genes. Unexpectedly, drug-induced hypo-phosphorylated Rb also bound thousands of non-promoter sites that gained H3K27ac upon CDK4/6 inhibition, and these regions were associated with transcriptional upregulation of nearby genes—many enriched for estrogen receptor targets. H3K27ac HiChIP showed that CDK4/6 inhibition caused Rb to localize to enhancer-rich regulatory hubs, many looping to ER target gene promoters. Notably, 60% of these Rb-associated hubs overlapped ER-bound loci. Canonical ER target promoters (e.g., CCND1, TFF1) and luminal markers (e.g., KRT18, KRT19) were loop-connected to Rb-bound enhancers and transcriptionally induced following treatment, suggesting that hypo-phosphorylated Rb directly engages enhancers to promote ER-driven transcription. Consistent with this, bulk RNA-seq revealed that estrogen response gene sets were among the top upregulated pathways in cell lines and PDX tumors after CDK4/6 inhibition. Transcriptomic analysis from the POP trial (N=44) showed that short-term palbociclib monotherapy induced consistent upregulation of estrogen response gene sets, supporting the clinical relevance of Rb-mediated transcriptional activation. Functional validation using CRISPR-Cas9 deletion of an Rb-bound enhancer looping to the KRT18 promoter impaired its induction by abemaciclib. RB1 knockout suppressed estrogen response gene upregulation genome-wide, confirming functional dependence on Rb. To further understand Rb-ER interaction, we studied isogenic MCF7-derived models harboring wild-type or mutant ESR1. In the wild-type setting, Rb activation by CDK4/6 inhibitors enhanced ER-responsive gene expression and reinforced luminal identity, explaining synergy with endocrine therapy. In ESR1-mutant cells, which exhibit ligand-independent ER activity, the Rb-associated enhancer program persisted despite estrogen deprivation (mimicking aromatase inhibition). This sustained expression of estrogen-responsive genes (including CCND1, encoding cyclin D1) offers a potential explanation for resistance to CDK4/6 inhibitor plus AI combinations in ESR1-mutant tumors. Finally, immunoprecipitation of Rb from nuclear lysates followed by mass spectrometry confirmed an interaction between Rb and ER after CDK4/6 inhibition, along with increased association of Rb with chromatin modifiers KDM5A and KDM5B. Knockdown of either KDM5A or KDM5B suppressed CDK4/6 inhibitor-induced, Rb-dependent estrogen response gene expression, implicating these demethylases as functional mediators linking hypo-phosphorylated Rb to ER-driven transcription. These findings redefine Rb as a dual-function regulator: repressing E2F-driven cell cycle genes while promoting estrogen-responsive transcription via enhancer engagement. This paradigm reshapes our understanding of Rb biology and highlights a mechanistic basis for CDK4/6 inhibitor activity, synergy with endocrine therapy, and resistance in ESR1-mutant tumors.

L. J. Esserman¹, A. S. Fiscalini¹, A. Naeim², L. J. van ‘t Veer¹, A. Kaster³, M. T. Scheuner¹, A. Z. LaCroix⁴, A. D. Borowsky¹, H. Anton-Culver⁵, O. I. Olopade⁶, J. N. Esserman⁷, R. Lancaster⁸, Y. Shieh⁹, E. Ziv¹, J. A. Tice¹, L. Madlensky⁴, A. Blanco¹, K. S. Ross¹, D. L. Goodman⁵, H. L. Park⁵, R. A. Hiatt¹, N. Wenger², B. A. Parker⁴, D. M. Heditsian¹, S. A. Brain¹, V. Lee¹, K. F. Rhoads¹⁰, K. Fergus¹, K. Blum¹, L. P. Sabacan¹, M. Eklund¹¹; ¹UCSF, San Francisco, CA, ²University of California, Los Angeles, Los Angeles, CA, ³Sanford Health, Fargo, ND, ⁴University of California, San Diego, San Diego, CA, ⁵University of California, Irvine, Irvine, CA, ⁶UChicago Medicine, Chicago, IL, ⁷Diagnostic Center of Miami, Miami, FL, ⁸University of Alabama at Birmingham, Birmingham, AL, ⁹Weill Cornell Medicine, New York, NY, ¹⁰Virginia Commonwealth University, Richmond, VA, ¹¹Karolinska Institutet, Stockholm, Sweden

Background: The goal of breast cancer (BC) screening is to reduce morbidity and mortality through the reduction of late-stage (Stage 2B or higher) cancers. Individual risk of BC varies substantially, and tailoring screening and prevention to individual risk could improve the focus on high risk women, reduce the burden of screening in low risk women and better allocate resources. Women Informed to Screen Depending On Measures of risk (WISDOM 1.0) Study was designed to test the non-inferiority of risk-based vs. annual screening for the primary outcome of Stage ≥2B BC, and to determine if risk-based screening is less morbid, preferred by women, and more conducive to prevention interventions.Methods: WISDOM 1.0, a randomized, preference sensitive pragmatic trial, compared the safety and morbidity of annual mammography versus risk-based screening, in which a woman’s BC risk is used to guide mammography initiation age, frequency, and the use of supplemental imaging and preventive interventions. Risk assessment included genetic testing (9 BC genes + polygenic risk scores (PRS)) and the Breast Cancer Surveillance Consortium (BCSC) v2 model, integrating breast density, to stratify women into four risk categories: highest (5-year risk >6%)/yearly mammography and MRI alternating every 6 months; elevated (top 2.5% of risk by age)/yearly mammography; average /biennial mammography; and lowest risk (no screening until 5-yr risk ≥1.3% or age 50). Risk-based participants had access to Breast Health Decisions, an educational tool to explain risk and options for risk reduction. Breast Health Specialists and genetic counselors provided consultations for women in the top 2.5% of risk by age or with pathogenic variants. Cancers, biopsies, chemoprevention, and imaging data were self-reported; 95% of all cancers were verified with medical records. Results: Over 7 years, 46,000 women enrolled, 61% chose to be randomized. Those who declined to randomize could elect to self-select their arm; 89% chose risk-based, regardless of age or geography. In the randomized cohort 77%, 6%, 4%, and 9% were White, African American, Asian, and Latina, respectively. There were 880 new BC diagnoses, of which 82 were Stage 2B or higher. Stage 2B cancer rate was non-inferior (p<0.001) and lower in the risk-based arm (42 vs. 28 stage ≥ IIB cancer per 100,000 person years in annual vs. risk-based, respectively, (p=0.15) for superiority). The proportion of participants in the randomized cohort assigned to the highest, elevated, average, and lowest risk was 2.1%, 8.0%, 63.2%, and 26.7%, respectively. Mammogram rates were lower in the risk-based arm (p<0.001). There was a trend towards more biopsies in the risk-based arm (p=0.08), driven by more biopsies in the highest two risk groups. The rates of invasive cancer, DCIS, mammograms, and biopsies varied by risk category, and were markedly higher in the high and elevated compared to average and lowest risk-based assignments (invasive cancer rates were 1279, 428, 233, and 169 per 100,000 women per year in the highest, elevated, average and lowest risk groups, respectively). Rates of chemoprevention increased among high risk groups in the risk-based arm from baseline to subsequent years (p<0.001). In women with pathogenic variants in BC susceptibility genes, 30% reported no family history of BC. Conclusions: A risk-based approach to BC screening and prevention is safe and acceptable to women and is an opportunity to improve breast cancer early detection and prevention by identifying the highest risk women who should be screened more frequently and offered risk reduction options while reducing screening burden for average risk women. WISDOM 2.0 is enrolling women 30-74 and is designed to identify younger women at high risk and improve prediction of fast and slow growing cancers using genetics and mammographic AI.

K. L. Jhaveri¹, P. Neven², M. Casalnuovo³, S. Kim⁴, E. Tokunaga⁵, P. Aftimos⁶, C. Saura⁷, J. O’Shaughnessy⁸, N. Harbeck⁹, L. A. Carey¹⁰, G. Curigliano¹¹, J. Watanabe¹², E. Lim¹³, J. Huang¹⁴, Z. Qingyuan¹⁵, A. Llombart-Cussac¹⁶, C. Huang¹⁷, B. Desai¹⁸, X. Wang¹⁸, S. Cao¹⁸, F. Bidard¹⁹, Y. Limay¹⁸; ¹Memorial Sloan Kettering Cancer Center, New York, NY, ²University Hospitals Leuven, Louvain, BELGIUM, ³Hospital Maria Curie, Buenos Aires, ARGENTINA, ⁴Asan Medical Center, University of Ulsan College of Medicine, Seoul, KOREA, REPUBLIC OF, ⁵Department of Breast Oncology, National Hospital Organization Kyushu Cancer Center, Fukuoka, JAPAN, ⁶Institut Jules Bordet, Brussels, BELGIUM, ⁷Vall d’Hebron University Hospital, Vall d’Hebron Institute of Oncology (VHIO), Barcelona, SPAIN, ⁸Baylor University Medical Center, Texas Oncology, US Oncology, Dallas, TX, ⁹Breast Center, Dept. OB&GYN; LMU University Hospital, Munich, GERMANY, ¹⁰University of North Carolina at Chapel Hill, Chapel Hill, NC, ¹¹University of Milano, Milan, Italy and European Institute of Oncology, IRCCS, Milan, ITALY, ¹²Juntendo University Graduate School of Medicine, Bunkyo, Tokyo, JAPAN, ¹³Garvan Institute of Medical Research, University of New South Wales, Sydney, AUSTRALIA, ¹⁴Xiangya Hospital Central South University, Changsha, Hunan, CHINA, ¹⁵Harbin Medical University Cancer Hospital, Harbin, CHINA, ¹⁶Hospital Arnau Villanova, Universidad CEU Cardenal-Herrera, Valencia, SPAIN, ¹⁷National Taiwan University Hospital and National Taiwan University College of Medicine, Taipei, TAIWAN, ¹⁸Eli Lilly and Company, Indianapolis, IN, ¹⁹Institut Curie, Paris, FRANCE.

Background: At the primary progression-free survival (PFS) analysis, the phase 3 EMBER-3 trial in patients (pts) with ER+, HER2- ABC demonstrated significant PFS benefit with imlunestrant (imlu) vs standard therapy (SOC: fulvestrant or exemestane) in pts with ESR1 mutations (ESR1m), and with imlunestrant+abemaciclib (imlu+abema) vs imlu in all pts, regardless of ESR1m status. Overall survival (OS) was immature. Here, we present updated efficacy from a prespecified interim OS analysis. Patients and Methods: Pts with ER+, HER2- ABC previously treated with aromatase inhibitors ± CDK4/6 inhibitors were randomized (1:1:1) to imlu, SOC, or imlu+abema. Primary endpoints were investigator-assessed PFS of imlu vs SOC in pts with ESR1m and all pts and of imlu+abema vs imlu in all concurrently randomized pts. OS was a key secondary endpoint (tested if the corresponding PFS was statistically significant). Due to only 2 of 3 PFS endpoints being met, limited significance level was passed to the OS comparisons. Exploratory endpoints included time to chemotherapy (TTC), PFS2, and efficacy analyses of imlu+abema vs SOC. Results: Between Oct2021 and Nov2023, 874 pts were randomly assigned 1:1:1 (imlu, n=331; SOC, n=330; imlu+abema, n=213). At data cut-off (18Aug2025), with a median follow-up of 28.5 months, 10.1% of pts remained on treatment (imlu, 10%; SOC, 5%; imlu+abema, 18%). In pts with ESR1m, the median OS (mOS) was 34.5 months for imlu vs 23.1 months for SOC (HR=0.60; 95% CI 0.43-0.86; p = 0.0043, boundary for significance not achieved). In all pts regardless of ESR1m, the mOS was not reached with imlu+abema vs 34.4 months with imlu (HR= 0.82, 95% CI: 0.59-1.16; p=0.2622). Updated PFS demonstrated sustained benefit from prior report (Table). Notably, in all pts regardless of ESR1m, the mPFS of imlu+abema vs imlu was 10.9 months vs 5.5 months (HR=0.59; 95% CI 0.47-0.74; nominal p<0.0001). All pre-specified exploratory endpoints favored imlu-based regimens. Safety profiles remain consistent with prior reports. Conclusions: At a median follow-up of 28.5 months, a clinically meaningful improvement in OS was observed with imlu vs SOC in pts with ESR1m (corresponding to a numeric increase in mOS of 11.4 months), however, the boundary for significance was not achieved. Also, a favorable OS trend emerged with imlu+abema vs imlu in all pts, regardless of ESR1m. Sustained benefit in PFS, with clinically meaningful improvement in TTC, and PFS2 further highlight the efficacy of imlu-based regimens. Taken together, these updated data reinforce the potential of imlu, as monotherapy or in combination with abema, as an all-oral chemotherapy-free targeted therapy option for ET-pretreated pts with ER+, HER2- ABC.

^a All concurrently randomized patients. Imlu + abema compared to ^b Imlu, ^c SOC ^d Not significant due to the two-sided significance level of 0.0000055 for this endpoint.

H. S. Rugo¹, S. M. Tolaney², K. L. Jhaveri³, M. Martin⁴, G. A. Vidal⁵, L. Moscetti⁶, A. Brufsky⁷, W. J. Gradishar⁸, A. Schneeweiss⁹, N. Niikura¹⁰, A. Favret¹¹, M. Alfie¹², K. S. Lee¹³, S. Khan¹⁴, M. Feldman¹⁵, B. M. Day¹⁵, L. H. Lam¹⁵, W. C. Darbonne¹⁵, T. M. Fernando¹⁵, P. Perez-Moreno¹⁵, E. L. Mayer¹⁶; ¹City of Hope Comprehensive Cancer Center, Duarte, CA, ²Dana-Farber Cancer Institute, Boston, MA, ³Memorial Sloan-Kettering Cancer Centre, New York, NY, ⁴Universidad Complutense, GEICAM, CIBERONC, Madrid, Spain, ⁵The West Cancer Center, Germantown, TN, ⁶Universitaria Policlinico Modena, Modena, Italy, ⁷University of Pittsburgh Cancer Institution, Pittsburgh, PA, ⁸Northwestern University, Chicago, IL, ⁹Heidelberg University Hospital and German Cancer Research Center, Heidelberg, Germany, ¹⁰Tokai University School of Medicine, Kanagawa, Japan, ¹¹Virginia Cancer Specialists, PC, Fairfax, VA, ¹²Organizacion Medica de Investigacion, Buenos Aires, Argentina, ¹³National Cancer Center, Goyang, Korea, Republic of, ¹⁴Nottingham University Hospitals NHS Trust, Nottingham, United Kingdom, ¹⁵Genentech, Inc., South San Francisco, CA, ¹⁶Dana-Farber Cancer Instituter, Boston, MA

Background The first-line standard of care (SOC) for patients (pts) with estrogen receptor-positive, HER2-negative advanced breast cancer (ER+, HER2- aBC) is CDK4/6 inhibitor (i) + endocrine therapy (ET) but effective post-CDK4/6i options remain limited. Giredestrant (GIRE) targets the ER pathway while everolimus (E) targets the PI3K/AKT/mTOR pathway; both of which are implicated in driving resistance in the post-CDK4/6i setting. evERA BC (NCT05306340) is the first Phase III trial to demonstrate statistically significant and clinically meaningful improvement in investigator-assessed progression-free survival (INV-PFS) with an all-oral selective ER antagonist and degrader combination of GIRE + E vs SOC ET + E in pts with ER+, HER2- aBC post-CDK4/6i + ET, both in pts whose tumors had a detectable ESR1 mutation (m) and in the intent-to-treat (ITT) population (Mayer ESMO 2025). The safety profile of GIRE + E was manageable with no unexpected findings (Mayer ESMO 2025). We report results from prespecified exploratory subgroup analyses. Methods Pts who had ER+, HER2- aBC with disease progression (PD) post-CDK4/6i + ET in the aBC setting, or relapse during/after CDK4/6i + ET in the adjuvant setting, were randomized 1:1 to once-daily oral 30 mg GIRE + 10 mg E or SOC ET (exemestane, fulvestrant, or tamoxifen) + E until PD or unacceptable toxicity. Mutational status was determined using circulating tumor DNA at baseline. The co-primary endpoints were INV-PFS per Response Evaluation Criteria in Solid Tumors v1.1 in pts whose tumors had detectable ESR1m and in the ITT population. INV-PFS was assessed by subgroup analyses. Results Three-hundred-and-seventy-three pts were randomized; 183 pts were randomized to GIRE + E and 190 to SOC ET + E. A total of 207 pts (55%) had tumors with ESR1m, 115 (31%) had PIK3CAm, and 137 (37%) had alterations (alt) in the PI3K pathway genes (PIK3CA/AKT1/PTEN). Sixty-four pts (17%) had both ESR1m and PIK3CAm; 76 (20%) had both ESR1m and PIK3CA/AKT1/PTEN alt. Ninety-eight percent of pts received a CDK4/6i in the metastatic setting. INV-PFS benefit was observed for GIRE + E vs SOC ET + E regardless of PIK3CAm status or PIK3CA/AKT1/PTEN alt in both the ESR1m and ITT populations (Table). Consistent benefit was observed regardless of duration of prior CDK4/6i (Table). Data for additional subgroup analyses for prior therapy will be presented. Conclusions GIRE + E led to clinically meaningful improvements in INV-PFS vs SOC ET + E irrespective of PIK3CAm and PIK3CA/AKT1/PTEN alt, and other key relevant subgroups, regardless of ESR1m status. These data support the use of GIRE + E vs SOC ET + E in pts with ER+, HER2- aBC post-CDK4/6i + ET.